That is the uncomfortable part of D&I. Hiring targets can look good on paper. Training sessions can be completed. Employee resource groups can be active. Leaders can speak about inclusion. None of that tells HR whether the workplace is actually changing.
This lies in the data. D&I metrics provide HR professionals with an opportunity to go beyond activities and measure what is actually happening in terms of representation, pay, promotion, retention and the employee experience. This is critical as a program must yield some kind of change over time. When it does not, then it becomes imperative for the organization to analyze the cause of this problem and ways to address this. It forms the core of the discussion in this paper.
What Are D&I Metrics and Why Do They Matter?
Diversity & Inclusion (D&I) metrics include quantifiable pieces of information which are used by HR managers to evaluate the demographic makeup of the workforce, equality in organization processes and attitudes of employees.
It looks quite straightforward but there is something that needs to be considered while talking about diversity and inclusion.
Diversity involves people in the organization and includes aspects such as workforce demographic, hiring and inclusion of people in management and leadership. Inclusion on the other hand involves the experience of these people in the organization after they are hired and included in the organization.
That is why one workforce number cannot tell the full story.
A company may have a diverse hiring pipeline and still struggle to move people from underrepresented groups into management. It may also have good representation at senior levels while employees from those groups report a poor workplace experience.
This is where D&I metrics become useful. They give HR a way to connect the dots.
They also create a check against performative activity. A training programme, an annual awareness campaign or a new ERG may show that the company is taking action. But activity alone does not prove impact.
The better test is what changed after the action was taken.
Did representation improve? Did people move through the organization at similar rates? Did pay gaps narrow? Did retention change? Did employees report a better experience?
Those are the questions the data should help answer.
Tracking Workforce Representation and Demographics
Let’s begin with the most obvious aspect of D&I first. Who works in the organization?
It would be wise for HR staff to monitor workforce diversity at various stages and levels of the hierarchy, rather than grouping everything into one big percentage figure. Everything could seem fine until diversity suddenly becomes poor when employees rise to higher positions.
One example is provided by the WEF’s 2026 Global Gender Gap report, which states that in 57 economies, women’s presence decreases from 46.0% to 23.0% from entry-level positions to the C-suite.
That is a major difference.
It also changes how HR should think about representation. The question is not only whether a company hires diverse talent. It is whether that talent continues to move through the organization.
Applicant Pipeline Diversity
Begin with the hiring funnel.
HR should examine the demographic profile of those who apply, are screened, interviewed, offered and ultimately hired. Just looking at the final number who are hired can mask where a dropoff in diversity begins.
For instance, a firm might get applicants from a fairly diverse group. If that mix changes sharply during screening, HR has a reason to look at the criteria being used at that stage.
The same applies later in the process. A drop at the interview stage points to a different issue from a drop after offers are made.
This makes applicant pipeline diversity more useful than simply reporting how many diverse candidates were hired. It helps HR find the point where the pipeline changes.
Leadership Representation
The next question is what happens after people enter the company.
The ILO reported that women represented 40.1% of global employment in 2025 but only 30.5% of managerial positions.
The gap shows why representation needs to be tracked by level. A company can have reasonable diversity across its workforce and still have a much less diverse management layer.
HR should therefore look at entry-level representation, management representation and executive representation separately. Over time, these numbers can reveal whether the organization has a hiring issue, an advancement issue or a retention issue.
That is a much more useful picture than one workforce-wide percentage.
Also Read: How to Design a Total Rewards Package for Employee Retention: A Strategic Guide for HR Leaders?
Measuring Equity Through Pay, Promotion and Retention
Representation tells us who is in the room. Equity asks what happens once they get there.
This is where diversity and inclusion measurement becomes more meaningful. It could be that two employees have the same responsibilities and similar job performance but very unequal access to remuneration and promotions.
Pay Equity and Compensation Parity
Pay is one of the clearest areas to examine.
The ILO reported in September 2026 that women earn around 18% less than men on average globally. It also found that the global gender pay gap had narrowed by only two percentage points over the past decade.
That does not mean every company has the same gap. It does mean HR needs a proper process for checking its own numbers.
It is important that when doing salary analysis, it does not focus just on the average salary of the employees. It is important that Human Resources look at the similarities among the jobs and look at aspects like position, level, experience, responsibility, and job scope.
The report issued by the OECD on the topic of pay transparency in 2026 shows the reason behind why this has become a tough job to do. Out of the total 38 OECD countries, 21 countries, that is 55% of the total member countries, have compelled private sector organizations to report their gender pay gap by 2026. The figure is expected to rise to 84%. There will be also equal-pay auditing requirements in 26 countries instead of 10.
The direction is clear. Pay equity is becoming something organizations are expected to measure and review, not simply discuss.
Promotion Rates by Demographic
Pay is only one part of equity.
Promotion data can reveal another problem. If one demographic group consistently advances more slowly than others, the company needs to understand why.
HR should compare the percentage of eligible employees who receive promotions across demographic groups. That is more useful than simply counting how many people from each group were promoted.
Now it is time to delve into the rationale of the numbers. Perhaps it’s the performance reviews. Or the manager recommendations. Or the availability of development opportunities, or the sponsorship they receive, or the criteria used to determine who is ready for the next level.
The point of the metric is not to label one group as a problem. It is to find where the process may be producing different outcomes.
Voluntary and Involuntary Turnover
Retention completes the puzzle.
The HR department should be able to differentiate between voluntary and involuntary turnover and separate the statistics into demographics. If there is a problem of high voluntary turnover among one demographic group, it should cause concern.
Why are people leaving? Are they seeing fewer opportunities to progress? Do they feel excluded from decisions? Are managers creating different experiences across teams?
Exit interviews can help, but they should not carry the entire burden. Employee surveys, manager feedback and other sentiment data can provide more context.
This is where the metrics start connecting. Representation tells HR who is there. Promotion data shows who is moving. Retention shows who stays.
Looking at only one of those numbers can lead to the wrong conclusion.
Assessing the Invisible Inclusion and Employee Sentiment
Some of the aspects of inclusion cannot be captured by a mere count of numbers.
There may be employees who are from minority groups and feel like their voices are not as valued. While there may be no difference in formal opportunities for such employees, they might have fewer chances of participation or development.
This is why employee sentiment is as important as other metrics of representation and equity.
eNPS by Cohort
eNPS can give HR a useful starting point, but the company-wide score should not be the end of the analysis.
HR should segment the results by demographic cohort. It is quite simple. A mean may seem perfectly fine despite some groups being in completely different situations.
Similar principles apply to wider employee surveys. Seek to find discrepancies among the questions regarding belonging, manager support, development opportunities, psychological safety, and involvement in the decision-making process.
Here is an interesting example from the OECD’s 2026 Diversity and Inclusion report. The survey of its 2025 staff had a 61% response rate, which was much higher than the 41% of 2022. Moreover, it had more than 500 open-text responses.
One could see that the level of confidence regarding equal recruiting opportunities and channels for reporting the inappropriate behavior was relatively high. Nevertheless, employees seemed less confident about development opportunities and involvement in organizational decision-making.
That is why the comments matter.
A score can show that sentiment is weaker. Employee feedback can help explain what sits behind it.
ERG Participation
Employee Resource Groups can also provide useful signals, but membership numbers alone should not be treated as proof that an ERG is successful.
HR can track membership, participation, budgets and activities. Those are useful operating measures. They are not impact measures.
The bigger question is what the group changes.
Does it help employees build networks? Does it surface workplace problems? Does it support career development? Does leadership use its feedback when policies are reviewed?
Those outcomes tell HR much more than the number of events held during the year.
The same principle applies to D&I metrics more broadly. Counting activity is easy. Measuring whether that activity changed something is the harder part.
How HR Leaders Can Use Data to Identify Areas for Improvement
Collecting D&I data is not the difficult part anymore. Knowing what to do with it is.
Firstly, HR professionals have to figure out where they are in order to be able to set up their goals concerning some of the measures such as diversity ratio in leadership positions, promotions process, wage gap, or employee engagement.
Secondly, HR professionals have to stop focusing on vanity metrics. Ten workshops on diversity don’t mean the same thing as diversity itself. Increased numbers of ERG members are not necessarily connected with better chances for advancement in one’s career.
Accountability of leadership is also important. D&I should not be just another part of HR reporting that is occasionally checked by company leaders once a quarter. Those who are responsible for hiring, managing and promoting employees must have some incentives to look at those indicators as well.
Most importantly, HR should read the metrics together.
A retention problem might point back to inclusion. A leadership gap might begin with promotion decisions. A pay gap may expose weaknesses in job evaluation.
The useful insight often sits between the numbers.
Conclusion
D&I metrics are useful because they make uncomfortable questions harder to avoid.
A company can say its workforce is becoming more diverse. The data may show otherwise at the leadership level. It can say employees have equal opportunities. Promotion or pay data may tell a different story. It can point to high engagement across the business while a specific group feels left out.
It is for that reason that D&I metrics should not become yet another annual reportable activity.
HR executives must pay attention to where there are disparities in terms of representation and outcomes, and how the employee experience differs from what leadership thinks it is.
The best time to audit that measurement system is before the next D&I initiative begins, not after another year of activity has passed.
